目的 評(píng)價(jià)采用機(jī)器學(xué)習(xí)（ML）算法構(gòu)建的藥物相互作用（DDI）風(fēng)險(xiǎn)預(yù)測(cè)模型性能，為臨床應(yīng)用與科研提供參考。方法 檢索PubMed、Embase、中國(guó)學(xué)術(shù)期刊全文數(shù)據(jù)庫(kù)（CNKI）、萬(wàn)方數(shù)據(jù)知識(shí)服務(wù)平臺(tái)、維普數(shù)據(jù)庫(kù)（VIP），收集截至2023年1月31日的相關(guān)文獻(xiàn)，并使用PROBAST工具評(píng)估模型質(zhì)量。結(jié)果 共納入54個(gè)DDI預(yù)測(cè)模型，主要使用Drugbank、Twosides數(shù)據(jù)庫(kù)構(gòu)建模型，共涉及21種ML算法，以圖神經(jīng)網(wǎng)絡(luò)和深度神經(jīng)網(wǎng)絡(luò)使用最頻繁。藥物結(jié)構(gòu)特征是最常用的預(yù)測(cè)因子，藥時(shí)曲線下面積（AUC）為0.83～0.99。所有模型存在較高的偏倚風(fēng)險(xiǎn)，主要源于信息偏倚和黑盒效應(yīng)，但整體適用性風(fēng)險(xiǎn)低。結(jié)論 基于ML構(gòu)建DDIs風(fēng)險(xiǎn)預(yù)測(cè)模型對(duì)臨床用藥有一定參考價(jià)值，但模型質(zhì)量亟待提高，未來(lái)應(yīng)開發(fā)更具可解釋性的模型并驗(yàn)證其臨床實(shí)用性。

Objective To evaluate the performance of drug-drug interactions (DDI) risk prediction models constructed by machine learning (ML) algorithm, and to provide reference for clinical application and scientific research. Methods PubMed, Embase, WanFang Data, CNKI and VIP databases were electronically searched to retrieve all ML studies on predicting DDI from inception to January 31st, 2023. PROBAST tool was used to evaluate model quality. Results A total of 54 DDI prediction models were included. The models were mainly constructed using Drugbank and Twosides databases, involving 21 ML algorithms. Figure neural network and deep neural network were used most frequently. Drug structure characteristics were the most commonly used predictors, the AUC range of 0.83 to 0.99. All models have a high risk of bias, mainly due to information bias and black box effect, but the overall suitability risk was low. Conclusions Building DDI risk prediction models based on ML has certain reference value for clinical drug use, but the quality of the model needs to be improved. In the future, more interpretable models should be developed and their clinical practicability should be verified.

R969.2

中國(guó)藥品監(jiān)督管理研究會(huì)研究課題-基于多元證據(jù)體探索中成藥安全性評(píng)價(jià)方法的研究(2024-Y-Y-006)；臨床研究和成果轉(zhuǎn)化能力提升試點(diǎn)項(xiàng)目–中藥制劑研發(fā)——治療胃輕癱中藥復(fù)方佛香散(DZMG-ZJXY-23002)